Six and Twelve Carbon Fluorocarbon Derivatives of Sulfur Hexafluoride

Feb 3, 2018 - obtained which did not depress the melting point of the. 135-136° melting picrate obtained by the enamine route. The filtrates from the...
0 downloads 0 Views 421KB Size
146-2

NOTES

VOL.

25

the solid thus obtained from 95% ethanol gave a picrate, Anal. Calcd. for C ~ ~ H Z I Nneut. O ~ : equiv. 211.3. Found: m.p. 135-136', as the major fraction. This picrate was sub- neut. equiv. 212.7. sequently shown to be the derivative of 6-methyl-6,7The aminoester mixture ( V and VIII) was dehydrogenated dihydro-1,5-pyrindine ( IIb). and decarbethoxylated in the manner described in the A n a l . Calcd. for C15H14N407: C, 49.72; H, 3.86. Found: preparation of X. The product obtained by this procedure C, 49.73; H , 3.93. was chromatographed on a preparative gas chromatographic Two grams of this picrate was converted to the free base units and the amines (IIa and IIb), which occurred as a ( I I b ) by treatment with concd. ammonium hydroxide. The single peak, were collected. This material was treated with free base (IIb) was dried over sodium hydroxide and dis- equimolar quantities of picric acid in 95y0 ethanol. The tilled giving a colorless liquid, b.p. 208' (751 mm.), ny melting point of the crude picrate mixture s-hich resulted 1.5239, d;: 1.0007. was 121.5-127.5'. This corresponds to an 80:20% mixture Anal. Calcd. for CSHIIX:neut. equiv. 133.2. Found: neut. of 135-136: 147-149" melting picrates according to a preequiv. 132.1. viously prepared melting point mixture diagram of the two A styphnate derivative of the amine (IIb) was made m.p. isomers. After four recrystallizations of a portion of this 1.56-1 - - - 57' . crude picrate mixture 146 mg. of picrate, m.p. 135-136', vas A n a l . Calcd. for C1t"4N.108: C, 47.62; H, 3.70. Found: obtained which did not depress the melting point of the C, 47.79: H. 3.72. 135-136" melting picrate obtained by the enamine route. Another picrate, m.p. 147-149", was obtained in smaller The filtrates from the recrystallizations were combined and amounts during the fractional recrystallization mentioned solvent removed under reduced pressure. After txelve above. This picrate was subsequently shown to be the de- recrystallizations of the crude picrate residue which rerivative of 5-methyl-6,7-dihydro-l,5-pyrindine (IIa). A mained, 24.3 mg. of picrate, m.p. 147-149', was obtained. mixed melting point with the petroleum base picrate pre- This picrate, m.p. 147-149", showed no melting point deviously isolated,' m.p. 147-149", showed no depression. pression on admixture with the picrate, m.p. 147-149", The infrared spectra of the petroleum base picrate (m.p. obtained by the enamine route. The filtrates from the last 147-149') and the synthetic base picrate were identical. recrystallizations were combined and solvent removed under Anal. Calcd. for CljHi4N107: C, 49.72; H, 3.86; N, 15.46. reduced pressure. The residual crude picrate, after drying, Found:C,49.88;H,4.08;S, 15.40. consisted of 106 mg. and melted a t 113-117'. Bccording to Conversion of 100 mg. of this picrate to the free base the melting point mixture diagram of the two isomeric (IIa) by treatment with ammonium hydroxide yielded a picrates this residue corresponded to a 45: 5554 mixture of liquid which, after drying and distillation, boiled a t 207-208' 135-136': 147-149" melting picrates. (746mm.). DEPARTMEKT OF CHEMISTRY A styphnate derivative of IIa was prepared by adding about THEUNIVERSITY OF TEXAS 30 mg. of the amine (Ha) to about 1 cc. of a saturated soluAUSTIN, TEX. tion of styphnic acid in 957Gethanol. Recrystallization from 95yGethanol yielded small yellow needles, m.p. 172-173.5". (8) See ref. 1 for a description of the gas chromatographic A n a l . Calcd. for C15H14X40a: C, 47.6; H, 3.7. Found: C, apparatus. 47.5; H, 3.8. 6,7-Dihydro-l,5-pyrindzne (X) f r o m 4a-carbelhoxyoctahydro-1 ,5-pyrindine (IX). One-half gram of the aminoester (IX), prepared according to Albertson,' was slowly distilled Six and Twelve Carbon Fluorocarbon through 30% palladium-on-charcoal which was held between 350-360". As before, hydrogen was used as a carrier Derivatives of Sulfur Hexafluoride gas for the vapors. The reaction was complete in 48 hr. The product was converted to a picrate derivative and after two R. D. DRESDNER, T. M.REED111, recrystallizations from 9570 ethanol gave 0.28 g. (31%) of T. E. TAYLOR, A X D J. A. YOT-NO picrate, m.p. 180-181". This picrate showed no melting point depression on admixture with an authentic sample of Received February I , 1960 the picrate derivative of X. Alterntcte synthesis of 5- and 6-methyl-6,7-dihydro-115pyrindine (IIa and IIb). Dieckmann condensation of diElectrochemical reactions of organic sulfides ethyl &methyladipate afforded the P-ketoester mixture (I11 in anhydrous hydrogen fluoride have received some and VI) b.p. 117-119' (16 mm.); (lit..2 m.p. 107-108", 11-12 mm.). This mixture was monocyanoethylated by attention in recent years.' This work extends these adding 17 g. of acrylonitrile (0.32 mole) slowly with stirring investigations to include hexyl and phenyl sulfide. to 52 g. of the mixture of P-ketoesters (I11 and VI) (0.30 As in previous work, the compounds RfSFI, mole) in 60 g. of 1,4-dioxane containing 4.2 g. of Triton B. (Rf)&3F4,R,F and Rf-Rf (where R, is C6F13 or The temperature of the reaction mixture was maintained between 30-40" by external cooling during the addition. cyclo C6F11) were isolated or detected. The yields After 45 min. the addition was complete and the reaction of isolable product were lorn for phenyl sulfide. mixture was stirred an additional hour. After adding about Many of the above products that xere liquids at 40 cc. of diethyl ether to the reaction mixture it was washed 25" were purified by preparative scale chromatogbriefly Kith 5T0 hydrochloric acid, then once with water. After drying over sodium sulfate, the solvents were re(1) (a) -4.F. Clifford, H. K. El-Shamy, H. J. Emeleus, moved in oaezco and the residue distilled giving 48 g. of and R. T.Haszeldine, d . Chem. Soc. 2372 (1953). (b) W. colorless material (73%), b.p. 136" (0.5-1 mm.), ng 1.4642. A. Severson, T. J. Brice, and R. I. Coon, 128th Meeting Anal. Calcd. for CI2H,iOxX: C, 64.57; H, 7.62. Found: C, ACY, Minneapolis, Minn., Sept. 11-16, 1958. (c) R. Dresd64.58; H, 7.73. ner, J . Bm. Chem. Sac. 79, 69 (1957). (d) F. W. Hoffman, The monocyanoethylated product (IV and VII) obtained T. C. Simmons, R . B. Beck, H. V. Holler, T . Katz, R. J. above was then reductively cyclized as in the preparation Koshar, E. R. Larsen, J. E. llulvaney, F. E. Rogers, B. of octahydro-1,5-pyrindine.lAfter removal of Raney nickel Singleton, and R. S. Sparks, J . Am. Chew. SOC.79, 2424 and ethanol, the product was distilled giving a colorless (1957). (e) R. D. Dresdner and J. A. Young, J . Am. Chem. liquid, b.1). 151' (35mm.). SOC.81, 574 (1959). (f) J. A. Young and R. D. Dresdner, J . Org. Chem. 24, 1021 (1959). (9) R. N. Haszeldine and F. ( 7 ) S . F. Alhertson, J . Am. Chem. SOC.,72, 2594 (1950). Nyman, J . Chem. SOC.2684 (1956).

.-1UGUST

1960

SOTER

1-165

raphy in one-inch columns packed with hexade- m.w. 338, n,?:" 1.2501; reported4 for n-C6Flo, b.p. 57.32' cane on an inert support.* The products that were n*:' 1.2515. This material was analysed by chromat'ography using a 40-foot column with hexadecane on inert, acid solids at 2.5" were further purified by the method washed Celite; the main component appeared to present as of Tatlow3 using hot heptane for the recrystal- 7570 by weight while another peak representing 207'~ seemed to be isomeric and about 5Cr, was probably nonlizations. Freon 112, 1,2-difluoro-1,1,2,2-tetrachloroethane,isomeric impurity. An aliquot of the material was subjected prepa,rat,ivescale chromatographic purification in which was an excellent solvent for the solid products and to all but a small amount of the apparent isomeric impurity molecular weights were determined by the freezing was removed. The refractive index a t 25" of this purified point depression method in this medium. The molal material was 1.2498. The second largest fraction by weight amounted to 234 freezing point constant was determined using perg. and boiled between 117.8 and 118.5". On chromat.ographic fluorodicyclohexyl as the standard. Raoult's law was analysis this material demonstrat'ed a main peak of about observed a t concentrations up to 0.06 molal. Semi- 74% by weight and seven other impurity peaks. An aliquot quantitative purity estimations of all the products of this was purified chromatographically until all but about reported meredetermined by vapor phase chromatog- 1% of the impurities were removed. It had the correct raphy using in all cases for the liquids both a polar analysis for perfluorohexylsulfur pentafluoride, C6F18SFjJ calcd. MRn 41.28, mol. wt. 448; observed values: MRp substrate such as the ethyl ester of Kel-F acid 8114 41.77, n y 1.2829, dy 1.8910, b.p. 118.2", m.p. -31.5 to and a nonpolar substrate such as hexadecane. These -30.5", mol. wt. 446. were run a t various flows of nitrogen carrier gas and Anal. Calcd. for C B F ~ ~ SF, F ~76.60; : S, 7.18. Found: F, at temperatures between 80 and 100". Very 76.7; S, 6 A fractionation flat x a s encountered in the vacuum fracsignificant changes in purity (as determined by tionat,ion of the higher boiling materials which condensed t o the integrated areas under the various chroma- a slush. This occurred a t 65' and about 1 mm. The fraction tographic peaks) mere noted before and after pre- amounted t o 100 g. It as filtered at, 0". The residues began parative scale purifications on narrow boiling frac- to melt, a t 50'. Aft,er several recrystallizations from hot tions obtained by fractionation. In all cases only heptane the white crystals reached a constant melting point, of 71-72'. A4purified sample was analysed chromaminor impurity components, 1-2%, were observed tographically using a stationary phase of the highest boiling in the purified materials. resin recovered from the cell residiies. Only about a 2% Both compounds, hexyl and phenyl sulfide, impurity \vas detected under the conditions of the tests. yielded some extremely high boiling (210-240" Molecular weights of the compounds were determined. a t 0.1 mm.) resinous products. One such sample was About 50 g. of the material, which analysed correct'lp for perfluorodihexylsulphur tetrafluoride, was recovered, calcd. used to prepare a chromatographic column on mol. wt. 746; found mol. wt. 725, 730, m.p. i1-72". which the purity of the purified solid products could Anal. Calcd. for C,?F&: F, 76.4; S,4.30. Found: F, 76.8; be estimated. These products were dissolved in S, 4.41. At 8501 51 mm., a fract,ion amounting to 51 g. x a s obFreon 113 and injected into the unit operated a t tained that by analogy with other work should have been 200". Ci2F?6.It had a wide melting range beginning a t 10" and The electrochemical cell and operation pro- showed three equal peaks when it was analysed on the cedures were very similar to those described prev- fluorocarbon resin. Results fTom phenyl su1,fide. Four hundred grams of phenyl iouslyldIealthough minor alterations were made in a t 5.1 volts and about 10 the overhead reflux condenser to prevent the per- siilfide (2.1 moles) ele~t~rolysed amperes yielded 925 g. of products during 267 hr. of operafluorocyclohexane from clogging. Products ac- tion. The effluent gas product. was mainly perfluorocyclocumulated a t two points: in the cold traps a t the hexane, which amounted to 375 g. of crude product, puriend of the gas train and as the less volatile fluoro- fied by sublimat,ion.The crude mat,erialwas placed in a large carbon layer that accumulated a t the bottom of st,oppered glass filter funnel. As perfluorocyclohexane on t,he walls of t,he funnel the trapped liquid imthe cell. In each operation the cell-retained prod- condensed purities passed t,hrough the filter and were removed via ucts were siphoned from the cell into a poly- a stopcock. The process was repeated until no more liquid ethylene separatory funnel and separated from a was collected. The vapor above the subliming solid had a small layer of hydrogen fluoride. This fluoro- composition of 99.3y0 of one component, as analysed chrocarbon material was neutralized with dilute aqu- matographically. The purified material had a molecular xeight of 300. Three hundred grams was recovered. eous sodium bicarbonate, dried, filtered, and frac-4fraction amounting to 30 g. was recovered in the fractionated in appropriate equipment. tion that boiled between 109.5 and 110.5". Chromatographic EXPERIMESTAL

Results .from hexyl suljde. Four hundred and seventy (2.3 moles) was electrolysed a t 4.6 volts grams of (C~HIB)DS and a t about 10 amperes and yielded 1041 g. of products during 340 hr. of operation. The main product by weight, 411 g., wzts a mixture of C6F14 isomers, b.p. 57.2-58.0", (2) T. RZ. Reed, J. F. Walter, R. Cecil, and R . D. Dresdner, I n d . and Eng. Cham. 51, 271 (1959). (3) G. 13. Barlow and J. C. Tatlow, J. Chem. SOC.4695 (1952).

purification yielded a product in which t,he main component was 98.9yc. Analysis shomed this material to be perfluorocyclohexylsulphur pentafluoride, c-C6Fl1SFj, calcd., mol. wt. 405, M R D 39.08; observed mol. wt. 404, ny 1.3041, d:5 1.9530, RlRD 39.67, b.p. (micro) 110.5', m.p. glass. -4nal. Cslcd. for C6F168: F, 74.40; S,7.83. Found: F, 74.00; s, 7.57. A second fraction was isolated a t 85" and 35 mm. which amounted to 117 g. of a slush. The slush was filtered, washed with water and acetone, and dried. The crude material (4) 8 . M. Lovelace, W. Postelnek, and D. A. Rausch, Aliphatic Fluorine Compounds, Reinhold Publishing Corp., New York, 1958, page 74.

1466

NOTES

melted a t 65.2-67.2'. Chemical analysis showed only a trace of sulfur. The material was recrystallized from boiling heptane several times to a constant melting point of 75.576.0'. This agrees with the reported8 value for perfluorodicyclohexyl. This compound was analysed by vapor phase chromatography using the fluorocarbon resin column and appeared to be 99% one component. The infrared spectrum of a melt agreed exactly with that for perfluorodicyclohexyl as determined by Tat1ow.a The carbon and fluorine analyses of the compound were within 0.3% of the calculated values. -4final fraction was isolated a t 126" and 37 mm. This was a slush amounting to only 12 g. Six grams of crude filtration reeldue melted a t 6567.5'. It decomposed a t 180 to 200" and accordingly could not be analysed by vapor phase chromatography. At lower temperatures the development time was so long that only one broad low peak was obtained. Consequently, the material was recrystallized from hot heptane to a constant melting point of 90-91'. Analysis and molecular weight by the cryoscopic method in Freon 112 indicated that the purified material was the perfluorodicyclohexylsulphur tetrafluoride, ( CBF11)2SFI, mol. n t . 670; observed mol. wt. 660, m.p. 90-91'. .lnal. Calcd. for Ci?FesS. F, 73.73; S, 4.77. Found: F, 73.38; S, 4.81. In both of the reported operations considerable material boiling well above the highest boiling products reported was obtained. This is not uncommon in low voltage electrolyses in hydrogen fluoride. Better yields of the cyclic derivatives would probably have been obtained if the starting sulfide had been the cyclohexyl sulfide instead of the phenyl derivative. Further evidence for this statement is apparent from the fact that, in general, unsaturated compounds even though resonance stabilized tend to run poorly in the cells and frequently cause corrosion of the electrodes. Although the corrosion in this case was not as severe as that generally encountered with unsaturated materials, nickel bearing compounds were found in the cell in some quantity. Furthermore, they were observed to react with water quite violently with the evolrition of heat and sparks.

Acknowledgment. The authors wish to thank Professor J. C. Tatlow and Dr. R. Stephens, of the University of Birmingham, England, for an infrared spectrum of perfluorodicyclohexyl. DEPARTMENTS O F NUCLE.4R ENGINEERING A N D CHEMISTRY UNIVERSITY OF FLORIDA GAINESVILLE, FLA.

VOI..

25

limited solubilities in polar solvents or react with them. Studies show that Dowex-50 is effective in removing various aromatic amines from a solution of the amine in anhydrous 1,2-dimetho~yethane.~ The degree of swelling of the ion exchange resin is, as might be expected, a function not only of the solvent but also of the extent to which the resin is crosslinked. The rate of adsorption of the amine is a function of the steric requirements of the amine as well as the degree of swelling and hence the porosity of the resin. By selectively swelling the ion exchange resin it is possible to get a L'molecular sieve" effect where the rate and extent of adsorption is sharply dependent upon the steric requirements of the amine. The use of Dowex-50 in 1,2dimethoxyethane was decided upon as a result of an investigation of the degree of swelling of Dowex-50 in a number of anhydrous ethers. As can be seen from an examination of Table I, the resin is not only swollen to a rather high degree in 1,Bdimethoxyethane, but there is also a strong variation of the degree of swelling with the extent of divinylbenzene crosslinkage. TABLE I SWELLING O F DRYDOWEX-BO

I N THE

HYDROGEN FORM

Cross linkage

MLILLX.

(% Divinylbenzene)

4 4 4 4 8 8 8 8 12 12 12 12

Solvent Water CHIOCH~CH~OCHI CHI(OCH,CH~)~OCHI 1,4-Dioxane Rater CH30CH2CHzOCHa CHI(O C H ~ C H ~ ) ~ O C H I 1,4-Dioxane Water CH30CHzCHzOCH3 CHI( OCH2CH2)20CH3 1,bDioxane

Swelling, 70vol.

215 110 92 100 115 93 80 75 85 60 55 35

Time, hr.

0.25 0.5 1 24 0.25 2 2 24 0.25 2 2 48

The Interaction of Some Arylamines w i t h Dowex 50 in 1,2 -Dimethoxyethane

It is seen that the degree of swelling decreases with increasing crosslinkage in each case. The rate of swelling in dioxane is slow, requiring fortyGIL,BERT S.P A N S O N A N D ROBERTELLSWORTH eight hours for maximum swelling in the case of the 12% crosslinked resin. The degree of swelling Received February 3, 1960 for 1,Bdimethoxyethane and also dimethyl ether of diethylene glycol is appreciable and differs The use of ion exchange resins has been largely markedly for resins of different crosslinkage. limited to aqueous solutions or polar solvents. To Moreover the time required for attainment of a certain extent this is because of the generally maximum swelling is reasonable. higher degree of resin swelling than that which Table I1 summarizes the results of the batchoccurs in nonpolar solvents. It is not necessary for wise extraction of some arylamines by Dowex-50 a resin to swell in order to exhibit an appreciable preswollen in l,&dimethoxyethane. The amine ion exchange capacity. However, swelling helps uptake is stated as milliequivalents of amine adto accommodate larger exchanging particles (mole- sorbed per gram of resin (dry basis). cules or ions).' Many organic compounds show

-

(1) R. Kunin, G. W. Bodamer, Znd. Eng. Chem., 45,2577

(1953).

(2) Ansul Ether-121. All glycol ethers were supplied by Ansul Chemical Co., Marinette, Wis.